With respect to its increasing significance which has recently been ascribed to vegetal substances as regenerative sources of raw materials, one of the objects of biotechnological research is to try to adapt vegetal raw materials to the demands of the processing industry. In order to enable the use of modified regenerative raw materials in as many areas as possible, it is furthermore important to obtain a large variety of substances.
Apart from oils, fats and proteins, polysaccharides constitute the essential regenerative raw materials derived from plants. Apart from cellulose, starch maintains an important position among the polysaccharides, being one of the most significant storage substances in higher plants. Besides maize, rice and wheat, potato plays an important role as starch producer.
The polysaccharide starch is a polymer made up of chemically homogeneous basic components, namely the glucose molecules. However, it constitutes a highly complex mixture from various types of molecules which differ from each other in their degree of polymerization and in the degree of branching of the glucose chains. Therefore, starch is not a homogeneous raw material. One differentiates particularly between amylose-starch, a basically non-branched polymer made up of .alpha.-1,4-glycosidically branched glucose molecules, and amylopectin-starch which in turn is a complex mixture of various branched glucose chains. The branching results from additional .alpha.-1,6-glycosidic interlinkings. In plants which are typically used for starch production, such as, e.g., maize or potato, the synthesized starch consists of about 25% of amylose starch and of about 75% of amylopectin starch.
In order to enable as wide a use of starch as possible, it seems to be desirable that plants be provided which are capable of synthesizing modified starch which is particularly suitable for various uses. A possibility of providing such plants is--apart from breeding--in the specific genetic modification of the starch metabolism of starch-producing plants by means of recombinant DNA techniques. However, a prerequisite therefor is to identify and to characterize the enzymes involved in the starch synthesis and/or the starch modification as well as to isolate the respective DNA molecules encoding these enzymes.
The biochemical pathways which lead to the production of starch are basically known. The starch synthesis in plant cells takes place in the plastids. In photosynthetically active tissues these are the chloroplasts, in photosynthetically inactive, starch-storing tissues the amyloplasts.
The most important enzymes involved in starch synthesis are starch synthases as well as branching enzymes. In the case of starch synthases various isotypes are described which all catalyze a polymerization reaction by transferring a glucosyl residue of ADP-glucose to .alpha.-1,4-glucans. Branching enzymes catalyze the introduction of .alpha.-1,6 branchings into linear .alpha.-1,4-glucans.
Furthermore, it is discussed that other enzyme activities, such as hydrolytic or phosphorolytic activities, are involved in the synthesis of starch (Preiss in Oxford Survey of Plant Molecular and Cell Biology, Oxford University Press, Vol. 7 (1991), 59-114). It can furthermore not be precluded that the "R enzyme", or the so-called disproportionizing enzyme, and the starch phosphorylases also are involved in starch synthesis, although these enzymes so far have been connected with the degradation of starch.
Starch synthases may be divided up in two groups: the granule-bound starch synthases (GBSS), which are mainly present bound to starch granules but also in soluble form, and the soluble starch synthases (SSS). Within these classifications, various isotypes are described for various species of plants. These isotypes differ from each other in their dependency on primer molecules (so-called "primer dependent" (type II) and "primer independent" (type I) starch synthases).
So far only in the case of the isotype GBSS I its exact function during starch synthesis has been successfully determined. Plants in which this enzyme activity has been strongly or completely reduced, synthesize starch free of amylose (a so-called "waxy" starch) (Shure et al., Cell 35 (1983), 225-233; Visser et al., Mol. Gen. Genet. 225 (1991) , 289-296; WO 92/11376); therefore this enzyme has been assigned a decisive role in synthesizing amylose-starch. This phenomenon is also observed in the cells of the green alga Chlamydomonas reinhardtii (Delrue et al., J. Bacteriol. 174 (1992), 3612-3620). In the case of Chlamydomonas it was furthermore demonstrated that GBSS I is not only involved in the synthesis of amylose but also has a certain influence on amylopectin synthesis. In mutants which do not show any GBSS I activity a certain fraction of the normally synthesized amylopectin, exhibiting long chain glucans, is missing.
The functions of the other isotypes of the granule-bound starch synthases, particularly GBSS II, and of the soluble starch synthases are so far not clear. It is assumed that soluble starch synthases, together with branching enzymes, are involved in the synthesis of amylopectin (see, e.g., Ponstein et al., Plant Physiol. 92 (1990), 234-241) and that they play an important role in the regulation of starch synthesis rate.
For potato, the isotypes GBSS I, GBSS II, as well as two or three isotypes of the soluble starch synthases, which so far have not been characterized further, have been identified (Ponstein et al. Plant Physiol. 92 (1990), 234-241; Smith et al., Planta 182 (1990), 599-604; Hawker et al., Phytochemistry 11 (1972), 1287-1293). Also for pea a GBSS II could be found (Dry et al., The Plant Journal 2,2 (1992), 193-202).
A cDNA encoding GBSS I from potato as well as a genomic DNA have already been described (Visser et al., Plant Sci. 64 (1989), 185-192; van der Leij et al., Mol. Gen. Genet. 228 (1991), 240-248). So far, no nucleic acid sequences encoding further granule-bound starch synthases or one of the soluble starch synthase isotypes from potato, have been reported.
Soluble starch synthases have been identified in several other plant species apart from potato. Soluble starch synthases have for example been isolated in homogeneous form from pea (Denyer and Smith, Planta 186 (1992), 609-617) and maize (WO 94/09144). In the case of pea it was found that the isotype of the soluble starch synthase identified as SSS II is identical with the granule-bound starch synthase GBSS II (Denyer et al., Plant J. 4 (1993), 191-198). In the case of other plant species the existence of several SSS-isotypes was described by means of chromatographic methods, as for example in the case of barley (Tyynela and Schulman, Physiologia Plantarum 89 (1993) 835-841; Kreis, Planta 148 (1980), 412-416), maize (Pollock and Preiss, Arch. Biochem. Biophys. 204 (1980), 578-588) and wheat (Rijven, Plant Physiol. 81 (1986), 448-453). However, DNA sequences encoding these proteins have so far not been described.
A cDNA encoding a soluble starch synthase so far has only been described for rice (Baba et al., Plant Physiol. 103 (1993), 565-573).